Distinct decidualization and senomorphic responses in endometrial vs. menstrual blood stem/stromal cells: implications for reproductive therapies.

Endometrial biopsies were collected at days 20.18 ± 1.98 (17–23) of the luteal phase. The study used endometrial samples collected from 20 infertile women receiving care at Bahman Hospital’s IVF Center in Tehran, Iran. These patients were scheduled for IVF procedures primarily due to male-factor infertility or other causes unrelated to endometrial pathology. During hysteroscopic examination, biopsies were taken from the functional layer of the endometrium, which was confirmed to be free of pathological lesions by a skilled obstetrician/gynecologist and pathologist. Additionally, menstrual blood samples were obtained from 30 healthy volunteers with regular menstrual cycles, and at least one live birth. Following the exclusion of samples that were contaminated or technically compromised, 10 endometrial biopsies and 20 menstrual blood samples were deemed suitable for subsequent experiments. The inclusion and exclusion criteria for both sample types required participants to be between 20 and 40 years old, with no history of vaginal infections, autoimmune diseases, metabolic polycystic ovarian syndrome (PCOS), endometriosis, chemotherapy, or immunotherapy in the past 3 years. Participants also needed to test negative for HIV, HBV, HCV, and HPV and abstain from steroid or oral contraceptive use in the preceding 3 months. Informed consent was obtained from the participants before enrolment to the study. Endometrial biopsies and menstrual blood samples were transported to the laboratory under cold chain conditions in DMEM-F12 culture medium (Gibco, USA) supplemented with antibiotics (100 IU/mL penicillin G (Biochrom, Germany), 0.1 mg/mL streptomycin (Abbott, USA) and 0.25 μg/mL amphotericin B (Health Biotech, India)). After cell isolation, the cells were cultured with an optimized medium α-MEM (Gibco) containing GlutaMAX (Gibco) and 10 ng/mL bFGF (Sigma, USA), and expanded for further experimental use. All the experiments were performed using cell sources at early passages (P2-P5). All EnSCs were isolated and purified according to the protocol proposed by Barros et al. [ 28 ]. Briefly, the fresh biopsy tissues were minced to small fragments using a sterile scalpel. Then, the endometrial fragments were digested with 0.5 mg/mL collagenase I (Sigma) and 0.05 mg/mL deoxyribonuclease I (DNase I, Sigma) in DMEM-F12 culture media for 1 h at 37 °C. The mixture was stirred at 20-min intervals during this period. After the incubation time, the enzymatic activity was neutralized using equivalent volume of DMEM-F12 containing 10% fetal calf serum (FCS) (Gibco) and centrifuged at × 280 g for 5 min to recover the pellet. Then, the cell pellet was reconstituted and cultured in DMEM-F12 containing 10% FCS to leave adherent fibroblastic cells after 48 h. The adherent EnSCs were then expanded and frozen for further analyses. MenSCs were isolated following a previously published protocol [ 29 ]. In brief, menstrual blood was collected on the second day of the menstrual cycle using a Diva Cup (Lunette, Sweden) and processed as described. The samples were washed three times with cold culture medium, filtered through a 70 μm cell strainer, and cultured in DMEM-F12 medium (Gibco) supplemented with 10% FCS. After 48 h, adherent fibroblastic cells remained, which were then expanded and cryopreserved for subsequent analyses. Tri-lineage differentiation of the isolated MenSCs has already been investigated and approved by this research group [ 23 , 26 , 30 ]. The expression of MSC markers was determined by flow cytometry using PE-labeled mouse anti-human CD73 and CD105 (Beckman Coulter, cat no: B68176 and B92442 , respectively), and FITC-labeled mouse anti-human CD34, CD45 and CD90 (Beckman Coulter, cat no: IM1870, A07782 and IM1839U, respectively) antibodies. Fibroblastic nature of the cells was additionally confirmed using PE-conjugated mouse anti-human CD10 (BD Biosciences, catalog no. 555375). As a control, cells were stained with the appropriate isotype control antibodies. Samples were then analyzed using BD FACSLyric ™ flow cytometer and the FlowJo software v. 10 was used to analyze the data. EnSCs and MenSCs were first expanded and then seeded in DMEM/F12 medium with 10% charcoal-stripped (CS)-FCS (hormone-adsorbed FCS) at a density of 100,000 cells per well in a 24-well culture plate (SPL, Korea), with each well containing 500 µL of culture medium. Then, upon reaching 70% confluency, the cells were cultured overnight in phenol red-free DMEM-F12 (Sigma) supplemented with 2% CS-FCS for hormonal rest. After that, the cells were induced to decidualized with either PC protocol containing 0.5 mM 8-Bromoadenosine 3′,5′-cAMP (8-Br-cAMP, Thermofisher, USA) and 1uM medroxyprogestrone acetate (MPA, Sigma) for 6 days or EPC protocol containing 50 μM 8-Br-cAMP, 1uM MPA and 10 nM 17-β estradiol (Sigma) for 12 days. These protocols are being the most commonly used and widely accepted methods according to the literature [ 8 , 31 ]. The induction media in each protocol was changed every 3 days to refresh the components needed for decidualization and the collected cell culture supernatants at days 3 and 6 in PC and at days 3, 6, 9 and 12 in EPC were kept frozen at − 20 °C for future analyses. Decidualization is a dynamic and progressive process that takes place over a 6-day period in PC protocol, during which distinct changes occur in the proteome, metabolome, and inflammatory cytokine profile at different time points. To capture these temporally regulated events, we examined the early, middle, and late stages of the process, as they occur in vivo to capture important intermediate alterations during decidualization. Defining well-decidualized (WD) and poorly-decidualized (PD) sources was based on the median level of PRL secretion, since PRL is considered as a proxy indicator of decidualization in previous studies [ 32 ]: WD sources secreted PRL above the median, indicating a higher degree of decidualization, while PD sources secreted PRL below the median, reflecting a lower extent of decidualization. The changes in circularity index of EnSCs and MenSCs before and after decidualization were calculated according to the method published in current protocols [ 33 ]. Since cells adopt a circular morphology after decidualization, this protocol enables the quantification of EnSC circularity directly from transmitted light microscopy images, eliminating the need for time consuming and costly confirmatory experiments. In brief, cells in each image were manually segmented using ImageJ. A randomized selection of segmented cells from each image was then performed to minimize bias in the analysis. The circularity index—calculated as: 4π x area/perimeter 2 —was used to assess cell shape. All measurements were carried out using a relevant plugin in ImageJ version 1.54. Total RNA was isolated and cDNA was synthesized from 1 μg of RNA according to the instructions proposed by RNAJia kit for RNA isolation and RT-Roset for cDNA synthesis products from Roje technologies (Iran). The quantity and purity of total RNA were assessed using a NanoDrop-2000c spectrophotometer (Thermo Scientific, USA). Real time PCR analysis was done using applied biosystems stepone plus system. The resulted data from each gene was normalized to RPL19 as a housekeeping gene and the relative expression of target genes was determined. Primer sequences used were as follows: PRL , forward: AAGCTGTAGAGATTGAGGAGCAAAC and reverse: TCAGGATGAACCTGGCTGACTA; IGFBP1 , forward: CGAAGGCTCTCCATGTCACCA and reverse: TGTCTCCTGTGCCTTGGCTAAAC; RPL19 , forward: GCGGAAGGGTACAGCCAAT and reverse: GCAGCCGGCGCAAA. All conditioned media were collected from the cells every 3 days and cleared for any cell debris by centrifugation. The PRL human ELISA kit (Diazist, Iran) was used to measure the levels of PRL secreted in the cell supernatants. Measurement of IL-6 was also carried out according to the protocol suggested by manufacturer (Biolegend, USA). Amino acids were measured and analyzed by targeted LC–MS/MS analysis using QSight 210 MD (Perkin Elmer, USA). The hierarchical clustering and analyzing differentially expressed metabolites were done using Metaboanalyst version 6 [ 34 ]. Regarding glycolysis intermediates, glucose was determined by GLUC3 and LACT2 kits from Roche diagnostics, USA. Two protocols were used for treatment of EnSCs and MenSCs with senomorphics. In pre-treatment protocol, selected senomorphics were added to culture media the day after cell seeding for 5 days with media refreshment on day 4 and then the cells were subjected to hormonal rest for 2 days before decidualization induction. In treatment protocol, soon after seeding, the cells subjected to hormonal rest and the selected senomorphics were added upon starting the day of decidualization. The PC protocol was used to induce decidualization. The selected senomorphics were A83-01, a TGF-β receptor inhibitor, at 500 nM and 1uM (Sigma), rapamycin at 100 nM (Sigma), resveratrol at 12.5 μM (Cayman Chemical), sitagliptin at 10uM (Sigma), metformin at 1 mM (Sigma) and prednisolone at 0.5 μg/mL (Sigma). Senescence-associated β-galactosidase staining was carried out using a senescence β galactosidase staining kit (cell signaling, USA) as per the manufacturer’s guideline. After SAβG staining, cell culture wells were embedded in 1 μg/mL DAPI and two random images were captured under light or fluorescence (excitation 340–380 nm, emission 435–485 nm) microscopes (Olympus BX51 equipped with DP71 CCD camera) with 20 × objective and exposure set at 15 ms (for bright field images). Quantitative analysis of SAβG was conducted using an ImageJ plugin, following the algorithm detailed in the relevant paper [ 35 ]. The cell nuclei in DAPI images were counted using cell counter macro in ImageJ v. 1.54. To determine the significance of differences between two independent groups, the Mann–Whitney U test was used while the Wilcoxon signed-rank test was applied for paired data. The robust Z-score is calculated as ( xi  − median)/MAD, where MAD is the median absolute deviation from the median, providing an outlier-resistant measure of deviation from the central tendency for reliable classification. Correlations were evaluated using Spearman’s rank correlation coefficient. Violin plots were employed to display the full distribution of raw data (represented as dots), along with the median (central line) and the interquartile range (indicated by the lower and upper lines) for expression levels. P values less than 0.05 were interpreted as statistically significant. Statistical analyses were carried out using GraphPad Prim v. 8.4 (GraphPad Software Inc., La Jolla, CA, USA) and R v. 4.3.3.

The demographic and clinical characteristics of the donors are summarized in Supp. Table S1. Then, EnSCs and MenSCs were subsequently isolated, and the expression of MSC–associated surface markers was assessed in both cell types. EnSCs and MenSCs displayed high levels of CD73, and CD105 expression, while they were negative for the expression for CD34 and CD45. Although EnSCs exhibited a moderate level of CD90 expression, almost all MenSC sources showed a high expression level of this marker. Both EnSCs and MenSCs highly express CD10, supporting their shared stromal cell identity and origin (Fig.  1 A, B). Fig. 1 Comparing decidualization potential in EnSCs and MenSCs. Endometrial (EnSCs) and menstrual blood stromal cells (MenSCs) were isolated from endometrial biopsies and menstrual blood of women recruited to this study. Cells were cultured and in vitro decidualized with either PC or EPC protocol. In some settings, cell remained in culture for 8 days before induction of decidualization (extended culture), while in short term culture cells were decidualized 2 days after initial cell seeding. Cells were characterized by investigating the expression of mesenchymal and stromal markers ( A and B ) (red histograms correspond to isotype controls), assessment of cell morphology before and after decidualization with PC protocol ( C ), measurement of mean circularity ( D ), mean marginal circularity ( E ) (bars illustrate 95% confidence interval (CI), central dots represent the average and non-overlapping arrows depict significant difference between decidualized and undecidualized cells), and raw mean circularity in decidualized group ( F ), the expression level of PRL and IGFBP1 transcripts ( G ), measuring the concentration of PRL in cell culture supernatant of decidualized cells in short and extended culture approaches ( H ), comparing PRL secretion of well-decidualized (WD) and poor-decidualized (PD) cells in the extended culture ( I ) and comparing PRL secretion in PC and EPC decidualization protocols at different time intervals ( J ). Violin plots represent a full distribution of raw data (dots), median (central line), and interquartile range (lower and upper lines) for expression. Each dot is attributed to single cell source. P values less than 0.05, 0.01, 0.001 and 0.0001 are shown with *, **, ***, **** respectively. D Decidualized, uD Undecidualized Comparing decidualization potential in EnSCs and MenSCs. Endometrial (EnSCs) and menstrual blood stromal cells (MenSCs) were isolated from endometrial biopsies and menstrual blood of women recruited to this study. Cells were cultured and in vitro decidualized with either PC or EPC protocol. In some settings, cell remained in culture for 8 days before induction of decidualization (extended culture), while in short term culture cells were decidualized 2 days after initial cell seeding. Cells were characterized by investigating the expression of mesenchymal and stromal markers ( A and B ) (red histograms correspond to isotype controls), assessment of cell morphology before and after decidualization with PC protocol ( C ), measurement of mean circularity ( D ), mean marginal circularity ( E ) (bars illustrate 95% confidence interval (CI), central dots represent the average and non-overlapping arrows depict significant difference between decidualized and undecidualized cells), and raw mean circularity in decidualized group ( F ), the expression level of PRL and IGFBP1 transcripts ( G ), measuring the concentration of PRL in cell culture supernatant of decidualized cells in short and extended culture approaches ( H ), comparing PRL secretion of well-decidualized (WD) and poor-decidualized (PD) cells in the extended culture ( I ) and comparing PRL secretion in PC and EPC decidualization protocols at different time intervals ( J ). Violin plots represent a full distribution of raw data (dots), median (central line), and interquartile range (lower and upper lines) for expression. Each dot is attributed to single cell source. P values less than 0.05, 0.01, 0.001 and 0.0001 are shown with *, **, ***, **** respectively. D Decidualized, uD Undecidualized Although MenSCs are thought to derive from EnSCs, their collection during a distinct menstrual phase may lead to divergent characteristics. Importantly, MenSCs experience hormonal withdrawal, which could impair their decidualization capacity. In this regard, decidualization potency of these two cell types was assessed and compared. During decidualization, the fibroblast-like EnSCs and MenSCs differentiated to large polyhedral, polyploid cells (Fig.  1 C). To confirm decidualization, the circularity index was measured in EnSCs and MenSCs. Recent findings suggest that this index can effectively capture phenotypic changes related to decidualization without the need for complex experimental procedures [ 33 ]. Notably, in EnSCs and MenSCs, the density plot displayed two distinct peaks corresponding to decidualized and undecidualized conditions (Fig.  1 D). As shown in Fig.  1 E, the circularity index was significantly higher in decidualized EnSCs and MenSCs compared to their undecidualized counterparts, with non-overlapping arrows highlighting this significance. Additionally, the mean circularity in EnSCs exceeded that in MenSCs, reinforcing the greater decidualization potential of EnSCs (Fig.  1 F). To further validate decidualization in EnSCs and MenSCs, expression level of PRL and IGFBP1 transcripts was assessed and compared. Decidualization led to a 340-fold increase in PRL mRNA expression in EnSCs, compared to a 142.9-fold increase in MenSCs. IGFBP1 mRNA expression increased 14,972-fold and 34,540-fold in EnSCs and MenSCs, respectively upon decidualization (Fig.  1 G). These findings were further corroborated by measurement of PRL concentration in cell culture supernatants of decidualized cells (Fig.  1 H-I). This finding was validated in three different settings. In the first setting, MenSCs and EnSCs were subjected to decidualization under PC protocol with two approaches. In the first approach (short culture), cells were decidualized 2 days after initial cell seeding, while in the second approach (extended culture) cells were remained in the culture plate for 8 days before decidualization induction. Culture media were then collected on days 3 and 6 following decidualization. Although at days 3 and 6 of decidualization in short culture approach, some MenSC samples exhibited PRL secretion levels comparable or even higher than those of EnSCs, overally EnSCs produced higher PRL levels than MenSCs on days 3. In the extended approach, EnSCs were found to be superior in term of PRL production at both days 3 and 6 of decidualization period (Fig.  1 H). In the second setting, WD and PD EnSC and MenSC sources were compared for PRL secretion upon decidualization. To validate our stratification method, we performed an unbiased analysis using median absolute deviation (MAD) to calculate robust z-scores from PRL secretion data. This approach, applied without pre-defining groups, confirmed that positive z-scores correspond to WD sources and negative z-scores to PD sources (Supp. Fig S1), thereby validating our initial classification. At both days 3 and 6 of decidualization process, WD and PD EnSCs produced significantly higher PRL compared to their WD and PD MenSC counterparts. Indeed, while WD EnSCs produced higher PRL compared to PD EnSCs at both days 3 and 6, no statistical difference between WD and PD MenSC was observed (Fig.  1 I). Since both PC and EPC are standard protocols for inducing decidualization, and in order to compare the kinetics of decidualization between EnSCs and MenSCs, we assessed PRL secretion from both cell sources following induction with PC and EPC in the third experimental setting. It is known that the EPC protocol uses a lower concentration of cAMP than the PC protocol (50 µM in EPC vs. 0.5 mM in PC) and requires a longer differentiation period (12 days for EPC vs. 6 days for PC). Again, EnSCs were found to have greater capacity to produce PRL in both protocols. In EnSCs, PRL secretion peaked at day 9 under the EPC protocol, followed by a decline by day 12. However, in MenSCs, there was no PRL peak until day 12, with secretion levels gradually increased over time (Fig.  1 J). Decidualization is known to induce extensive molecular and metabolic reprogramming in EnSCs, including shifts in energy metabolism, amino acid utilization, and biosynthetic pathways that support embryo implantation and early pregnancy. However, these metabolomic alterations have not yet been characterized in MenSCs. In this study, WD and PD EnSCs and MenSCs were in vitro decidualized by PC protocol for 6 days and metabolome of cell culture supernatants was investigated (Supp. Data 1). Figure  2 A illustrates the heatmap of metabolome changes at days 3 and 6 of decidualization in EnSCs and MenSCs. The clustering of undecidualized and decidualized EnSCs, regardless of their decidualization capacity (WD and PD) occurred at day 6, while in MenSCs clustering was observed sooner at day 3. Notably, WD and PD sources exhibited differential clustering at both cell types. The heatmap plots showed that WD sources clustered at day 3 in EnSCs and day 6 in MenSCs. Figure  2 B shows that while Partial least squares discriminant analysis (PLS-DA) plots successfully separated undecidualized and decidualized cells into distinct clusters at both days 3 and 6, they do not differentiate between WD and PD sources in either EnSCs or MenSCs. The variable importance in projection (VIP) metabolites that contributed to clustering each group at each day are shown in Fig.  2 B. Notably, citrulline and tyrosine were VIP metabolites that shared among EnSCs on days 3 and 6 and MenSCs on day 3. Fig. 2 Metabolomic analysis of decidualization kinetics in EnSCs and MenSCs. A Hierarchical clustering analysis of metabolites with differential expression at 3 and 6 days after decidualization induction of EnSCs and MenSCs. Class “0” corresponds to decidualized cells, while class “1” represents undecidualized cells. The lower side of the panel A shows sample codes, and dendrogram on the left and top represents the clustering results of the differential metabolites. B Partial least squares discriminant analysis (PLS-DA) and Variable Importance in Projection (VIP) score plot for EnSCs and MenSCs at days 3 and 6 following decidualization. Each dot in PLS-DA plot represents single cell source. In VIP plot, metabolites with VIP scores > 1 are typically considered significant to separate groups. Red boxes indicate amino acids that are consistently shared across all the studied cells and time points. C Venn diagram of pathway enrichment analysis shows shared pathways between the studied cells at each day of decidualization. The color scale in heatmap and venn diagram represents the relative abundance of metabolites and pathways, respectively with red indicating higher abundance, deep blue indicating lower abundance, and light blue indicating zero abundance. The presented black asterisk defines shared pathways between EnSC days 3 and 6 and MenSC day 3. D Dot plot pathway enrichment map showing the significant enriched pathways and their direction of regulation in each cell at each decidualization timepoint. The positive and negative normalized enrichment score (NES) represent upregulation and downregulation of each pathway, respectively. E Violin plot of concentrations for significantly altered metabolites (p < 0.05) in decidualized (green) and undecidualized (red) cells. Y axes are represented as normalized units. Due to this normalization process we obtained negative scale in the Y-axis in some of the bins Metabolomic analysis of decidualization kinetics in EnSCs and MenSCs. A Hierarchical clustering analysis of metabolites with differential expression at 3 and 6 days after decidualization induction of EnSCs and MenSCs. Class “0” corresponds to decidualized cells, while class “1” represents undecidualized cells. The lower side of the panel A shows sample codes, and dendrogram on the left and top represents the clustering results of the differential metabolites. B Partial least squares discriminant analysis (PLS-DA) and Variable Importance in Projection (VIP) score plot for EnSCs and MenSCs at days 3 and 6 following decidualization. Each dot in PLS-DA plot represents single cell source. In VIP plot, metabolites with VIP scores > 1 are typically considered significant to separate groups. Red boxes indicate amino acids that are consistently shared across all the studied cells and time points. C Venn diagram of pathway enrichment analysis shows shared pathways between the studied cells at each day of decidualization. The color scale in heatmap and venn diagram represents the relative abundance of metabolites and pathways, respectively with red indicating higher abundance, deep blue indicating lower abundance, and light blue indicating zero abundance. The presented black asterisk defines shared pathways between EnSC days 3 and 6 and MenSC day 3. D Dot plot pathway enrichment map showing the significant enriched pathways and their direction of regulation in each cell at each decidualization timepoint. The positive and negative normalized enrichment score (NES) represent upregulation and downregulation of each pathway, respectively. E Violin plot of concentrations for significantly altered metabolites (p < 0.05) in decidualized (green) and undecidualized (red) cells. Y axes are represented as normalized units. Due to this normalization process we obtained negative scale in the Y-axis in some of the bins To gain insight on the potential connection between EnSCs and MenSCs during decidualization process, shared biological pathways at different time points of decidualization were investigated. Accordingly, the most commonly shared pathways occurred between decidualized EnSCs day 6 and MenSCs day 3, as well as between EnSCs day 3 and MenSC day 3, suggesting that day 6 decidualized MenSCs mostly have distinct pathways different from those used in day 3 decidualized MenSCs and day 3 and 6 decidualized EnSCs (Fig.  2 C). The significant pathways shared between EnSCs on days 3 and 6 and MenSCs on day 3 of decidualization included metabolism of phenylalanine, tyrosine, cysteine, methionine and creatinine, thyroxine production, all of which showed enrichment scores in the same direction (Fig.  2 D). Additionally, pathways associated with protein synthesis including total, cytosolic, and mitochondrial tRNA aminoacylation upregulated in EnSCs and MenSCs on day 3 but were downregulated in EnSCs by day 6 (Fig.  2 D). By analyzing differentially expressed amino acids, we observed that type and the direction of expression changes in some amino acids were also shared between the conditions. The increased level of phenylalanine, tyrosine and methionine were shared between EnSCs days 3 and 6 and MenSCs day 3. Glycine was decreased in both EnSCs day 6 and MenSCs day 3. However, the direction of expression change for alanine was opposite between MenSCs days 3 and 6 (Fig.  2 E). Univariate analysis was performed to identify the most effective predictors of the decidualization status in EnSCs, revealing that senescence, arginine, and citrulline were the top predictive markers, with area under the curve (AUC) values of 1, 1, and 0.98, respectively. Decidualization is a process of differentiation of EnSCs accompanied by dramatic changes in metabolic pathways including glucose metabolism. To this end, alteration of glucose and lactate levels during decidualization of EnSCs and MenSCs was assessed. We found that at day 3 of decidualization, no significant changes were observed in glucose and lactate levels in EnSCs; however, by day 6, decidualization led to a statistically significant increase in glucose and a corresponding decrease in lactate levels (Fig.  3 A). Interestingly, the pattern of glucose and lactate changes in EnSCs on day 6 resembled that of MenSCs on day 3. A significant decrease in lactate was also observed in MenSCs on day 6, although changes in glucose levels were not statistically significant (Fig.  3 B). To gain insight on potential impact of decidualization capacity on glucose metabolism, WD and PD EnSCs and MenSCs were compared. Akin to whole EnSCs sources, glucose and lactate changes in WD and PD EnSCs were not significant at day 3. At day 6, however, glucose concentration was increased in both WD and PD EnSCs, but lactate level on day 6 of decidualization was decreased only in WD EnSCs. Notably, glucose level in days 3 and 6 of decidualization was not affected by the decidualization capacity in MenSCs (Fig.  3 C). WD and PD MenSCs at day 3 and WD MenSCs at day 6 of decidualization showed decreased lactate level (Fig.  3 D). Fig. 3 Decidualization-associated changes in glucose metabolism of EnSCs and MenSCs. Glucose and lactate concentration in cell culture supernatant of uD and 3 and 6 days after decidualization in D cells in all sources of EnSCs ( A ) and MenSCs ( B ) , and after separation to well-decidualized (WD) and poor-decidualized (PD) in EnSCs ( C ) and in MenSCs ( D ). Each dot is attributed to single cell source. P values less than 0.05 and 0.01 are shown with *, ** respectively. D decidualized; uD undecidualized Decidualization-associated changes in glucose metabolism of EnSCs and MenSCs. Glucose and lactate concentration in cell culture supernatant of uD and 3 and 6 days after decidualization in D cells in all sources of EnSCs ( A ) and MenSCs ( B ) , and after separation to well-decidualized (WD) and poor-decidualized (PD) in EnSCs ( C ) and in MenSCs ( D ). Each dot is attributed to single cell source. P values less than 0.05 and 0.01 are shown with *, ** respectively. D decidualized; uD undecidualized Cellular senescence happens during decidualization and pro-senescent decidual response in the luteal phase endometrium is a prerequisite for induction of decidualization and PRL secretion. Accordingly, the level of cell senescence was investigated and compared in EnSCs and MenSCs. In vitro decidualization by PC protocol led to a significant increase in SABG activity in both EnSCs and MenSCs (Fig.  4 A). However, when cells were subjected to in vitro decidualization by a gentler induction method (EPC protocol), only MenSCs showed a significant increase in SABG activity (Supp. Fig. S2A). Notably, the baseline SABG activity was higher in MenSCs compared to EnSCs across both decidualization protocols (Fig.  4 A and Supp. Fig. S2A) reinforcing to the senescent nature of MenSCs. A comparison between WD and PD sources in two decidualization protocols further highlighted senescence differences in MenSCs and EnSCs. In in vitro decidualization by PC protocol, WD and PD MenSCs showed comparable level of senescence after decidualization, while PD EnSCs showed a significantly higher senescence than WD EnSCs (Fig.  4 B, C). Interestingly and in contrary to what observed in PC protocol, decidualization by EPC protocol resulted in a significant increase in SABG activity in PD MenSCs, while no statistical difference was found in SABG activity between PD and WD EnSCs (Fig.  4 C and Supp. Fig. S2C). Fig. 4 Assessment of cell senescence in EnSCs and MenSCs following decidualization. Cells were decidualized by PC protocol and SABG activity in undecidualized and decidualized EnSCs and MenSCs in ( A ) all sources were measured and compared. Panel ( B ) depict representative images of the cells after classification to well-decidualized (WD) and poor-decidualized (PD) groups. ( C ) The quantitative SABG activity was measured in WD and PD cell sources and ( D ) Level of IL-6 was measured in cell culture supernatant of undecidualized and decidualized EnSCs and MenSCs at different timepoints of decidualization. Each dot is attributed to single cell source. P values less than 0.05, 0.01, 0.001 and 0.0001 are shown with *, **, ***, **** respectively. IDV, Integrated density value Assessment of cell senescence in EnSCs and MenSCs following decidualization. Cells were decidualized by PC protocol and SABG activity in undecidualized and decidualized EnSCs and MenSCs in ( A ) all sources were measured and compared. Panel ( B ) depict representative images of the cells after classification to well-decidualized (WD) and poor-decidualized (PD) groups. ( C ) The quantitative SABG activity was measured in WD and PD cell sources and ( D ) Level of IL-6 was measured in cell culture supernatant of undecidualized and decidualized EnSCs and MenSCs at different timepoints of decidualization. Each dot is attributed to single cell source. P values less than 0.05, 0.01, 0.001 and 0.0001 are shown with *, **, ***, **** respectively. IDV, Integrated density value We next measured IL-6, as one of the most important components of SASP during decidualization of EnSCs and MenSCs. In line with the increased senescence observed after decidualization in EnSCs, IL-6 levels rose significantly in day 3 and 6 of decidualization. MenSCs also increased IL-6 secretion after decidualization on day 6, but no difference was observed at day 3. Supporting the earlier finding on SABG activity, MenSCs showed a higher baseline level of IL-6 secretion compared to EnSCs on both days 3 and 6 of decidualization (Fig.  4 D). When the cells were in vitro decidualized using EPC method, MenSCs produced significantly higher levels of IL-6 from day 6 to 12 of decidualization period compared to undecidualized cells, while such difference only was found on Day 9 of EnSCs decidualization (Supp. Fig. S2C). Correlation analyses were conducted to evaluate a potential link between PRL secretion, senescence, and IL-6. Notably, a negative correlation was identified between PRL secretion and senescence in all sources and WD EnSCs suggesting that the more senescent EnSCs are, the less they expected to produce PRL. No significant correlation, however, was observed in PD EnSCs. Additionally, a strong and significant negative correlation between PRL and IL-6 was observed in EnSCs, both across all sources and within WDs and PDs. In contrast, no similar significant correlations were found in MenSCs (Supp. Fig. S3). Detailed phenotypic and functional profiling of EnSCs vs. MenSCs has been provided in Supp. Table S2. Given the key role of senescence in decidualization, and the potential for different doses and timing of senomorphic application to either promote or suppress this process, we first examined their effects on the decidualization of both EnSCs and MenSCs across various timepoints and concentrations. According to the literature, senomorphics and the steroid prednisolone—which both influence senescence—were examined. These were chosen either because they are used for their potential effects on fertility or because they have been studied for their possible impact on aging. Initially, EnSCs and MenSCs were treated with these senomorphics using the concentrations and durations specified in previous studies [ 20 , 36 – 39 ]. Generally, treatment of the cells, particularly in EnSCs, with senomorphics during decidualization (treatment protocol) led to the inhibition of secretion of PRL (data not shown) and induction of senescent (see below). Based on these results, we chose pre-treatment approach for the subsequent experiments and selected one concentration per senomorphic for further experiments. Based on the initial set up (Fig.  5 A), a 5-day pretreatment protocol was developed for treating the cells with senomorphics. In all experiments, the standard PC protocol served as the control. In all sources of EnSCs as well as in WD and PD EnSCs, pretreatment with most senomorphics exerted no effect on PRL secretion except for metformin and prednisolone on day 3 and sitagliptin on day 6, which resulted in a significant decrease in PRL levels compared to the control (PC). Similar effects were observed in WD EnSCs at day 6, though the changes were close to being statistically significant (Fig.  5 B). In MenSCs, no significant changes were observed in most settings. However, by day 6, both in all sources and WD MenSCs, pretreatment with prednisolone resulted in a significant reduction of PRL level compared to PC control (Fig.  5 C). Fig. 5 Effect of pretreatment with senomorphics on PRL secretion by decidualized EnSCs and MenSCs. A Schematic representation of the protocol for pretreatment with senomorphics before the induction of decidualization. B Effect of senomorphics on PRL secretion by EnSCs and MenSCs. C . The left panels display PRL levels across all sources, unstratified, at days 3 and 6, while the middle and right panels present PRL levels specifically in WD and PD sources, respectively. Each dot is attributed to single cell source. P values less than 0.05, 0.01 are shown with *, ** respectively. WD Well-decidualized; PD Poor-decidualized; Rapa Rapamycin; Rez Resveratrol; Sita Sitagliptin; Met Metformin; Pred Prednisolone Effect of pretreatment with senomorphics on PRL secretion by decidualized EnSCs and MenSCs. A Schematic representation of the protocol for pretreatment with senomorphics before the induction of decidualization. B Effect of senomorphics on PRL secretion by EnSCs and MenSCs. C . The left panels display PRL levels across all sources, unstratified, at days 3 and 6, while the middle and right panels present PRL levels specifically in WD and PD sources, respectively. Each dot is attributed to single cell source. P values less than 0.05, 0.01 are shown with *, ** respectively. WD Well-decidualized; PD Poor-decidualized; Rapa Rapamycin; Rez Resveratrol; Sita Sitagliptin; Met Metformin; Pred Prednisolone Cellular senescence may play a dual role—contributing to normal decidualization while also potentially driving repeated implantation failure and recurrent pregnancy loss. Currently, research efforts are intensely focused on elucidating the biological pathways responsible for female reproductive aging. Emerging data increasingly implicates cellular senescence as a key player in this process. In this context, emerging senotherapies, aimed at mitigating or reversing reproductive aging, are being investigated to enhance implantation and pregnancy success [ 20 ]. Based on differential senescence of WD and PD sources, here we aimed to investigate to which extent decidualization capacity of these cells are affected by senomorphics treatment. Interestingly, pre-treatment of EnSCs with any senomorphic tested in this study profoundly increased SABG activity and induced senescence in WD EnSCs, whereas the opposite effect was observed in PD EnSCs (Fig.  6 A). Contrary to EnSCs, similar pretreatments did not exert any tangible effect on MenSCs senescence (Fig.  6 B). As noted in the previous section, senomorphic treatment during decidualization (treatment protocol) negatively impacted PRL levels in cell sources. To explore the potential effect on SABG, we assessed the impact of senomorphic treatment on both EnSCs and MenSCs during decidualization. Interestingly, this approach resulted in a significant increase in SABG activity and senescence in both EnSCs and MenSCs. Unlike the pretreatment protocol, which differentially affected WD and PD sources in EnSCs, treatment protocol resulted in a significant increase in SABG in both WD and PD EnSCs sources (Fig.  6 C, D). Of note, PD MenSCs showed no responsiveness to the treatment with senomorphics (Fig.  6 D). Fig. 6 Effect of senomorphics on senescence of EnSCs and MenSCs. Cells were either pretreated with senomorphics for 5 days before decidualization ( A and B ) or treated with them once decidualization is started ( C and D ) and SABG activity was measured. Each dot is attributed to single cell source. P values less than 0.05, 0.01, 0.001 and 0.0001 are shown with *, **, ***, **** respectively. IDV Integrated density value; WD well-decidualized; PD poor-decidualized; Rapa Rapamycin; Rez Resveratrol; Sita Sitagliptin; Met Metformin; Pred Prednisolone Effect of senomorphics on senescence of EnSCs and MenSCs. Cells were either pretreated with senomorphics for 5 days before decidualization ( A and B ) or treated with them once decidualization is started ( C and D ) and SABG activity was measured. Each dot is attributed to single cell source. P values less than 0.05, 0.01, 0.001 and 0.0001 are shown with *, **, ***, **** respectively. IDV Integrated density value; WD well-decidualized; PD poor-decidualized; Rapa Rapamycin; Rez Resveratrol; Sita Sitagliptin; Met Metformin; Pred Prednisolone The impact of pre-treatment with senomorphics on IL-6 production was assessed in the next step. In line with a significant resolution in SABG activity in PD EnSCs, pre-treatment with senomorphics caused reduced levels of IL-6 secretion on day 3. Interestingly and despite higher SABG activity in WD EnSCs following pre-treatment with senomorphics, IL-6 production by these cells was unaffected (Fig.  7 A). MenSCs differentially responded to pre-treatment with most senomorphics in that A8301, resveratrol, sitagliptin and metformin resulted a reduced levels of IL-6 section on day 6 in MenSCs in general and PD MenSCs in particular (Fig.  7 B). Fig. 7 Effect of senomorphic pretreatment on IL-6 level in EnSCs and MenSCs. Cells were pretreated with senomorphics for 5 days before decidualization and then in vitro decidualized. Concentration of IL-6 was determined at days 3 and 6 of decidualization in EnSCs ( A ) and MenSCs ( B ). Each dot is attributed to single cell source. P values less than 0.05, 0.01 are shown with *, ** respectively. WD Well-decidualized; PD Poor-decidualized; Rapa Rapamycin; Rez Resveratrol; Sita Sitagliptin; Met Metformin; Pred Prednisolone Effect of senomorphic pretreatment on IL-6 level in EnSCs and MenSCs. Cells were pretreated with senomorphics for 5 days before decidualization and then in vitro decidualized. Concentration of IL-6 was determined at days 3 and 6 of decidualization in EnSCs ( A ) and MenSCs ( B ). Each dot is attributed to single cell source. P values less than 0.05, 0.01 are shown with *, ** respectively. WD Well-decidualized; PD Poor-decidualized; Rapa Rapamycin; Rez Resveratrol; Sita Sitagliptin; Met Metformin; Pred Prednisolone We showed earlier that EnSCs and MenSCs have differential glucose metabolism during decidualization, which was more prominent at day 6. Indeed, decidualized WD and PD sources exhibited different glucose uptake and lactate production. In this regard, we investigated in the next step the potential impact of pre-treatment with senomorphics on glucose metabolism in these two-cell population. Although, glucose uptake by EnSCs and MenSCs was not affected in general, PD EnSCs showed diminished glucose consumption following pre-treatment with all senomorphics except A8301 (500 nm) at day 6 of decidualization (Fig.  8 A). In contrary to the diminished levels of lactate production during decidualization, pre-treatment of MenSCs with rapamycin and resveratrol caused higher production of lactate at day 6 of decidualization (Fig.  8 B). Nonetheless, in both EnSCs and MenSCs, lactate production by WD sources was unaffected by senomorphics, while PD sources showed a downward trend of lactate production after pre-treatment with most of the senomorphics. Fig. 8 Effect of senomorphic pretreatment on glucose metabolism in EnSCs and MenSCs. Cells were pretreated with senomorphics for 5 days before decidualization and then in vitro decidualized. Glucose ( A ) and lactate ( B ) levels in cell culture supernatant of EnSCs and MenSCs were them measured. Each dot is attributed to single cell source. P values less than 0.05, 0.01 are shown with *, ** respectively. WD Well-decidualized; PD poor-decidualized; Rapa Rapamycin; Rez Resveratrol; Sita Sitagliptin; Met Metformin; Pred Prednisolone Effect of senomorphic pretreatment on glucose metabolism in EnSCs and MenSCs. Cells were pretreated with senomorphics for 5 days before decidualization and then in vitro decidualized. Glucose ( A ) and lactate ( B ) levels in cell culture supernatant of EnSCs and MenSCs were them measured. Each dot is attributed to single cell source. P values less than 0.05, 0.01 are shown with *, ** respectively. WD Well-decidualized; PD poor-decidualized; Rapa Rapamycin; Rez Resveratrol; Sita Sitagliptin; Met Metformin; Pred Prednisolone According to the results presented above, we observed that, in contrast to WD sources, pre-treatment of PD EnSCs improved the cell senescence, caused diminished level of IL-6 production and exerted a positive impact on glucose metabolism. To further delineate these observations, we investigated whether pre-treatment with senomorphics could have a positive impact on metabolomics of PD EnSCs. Our results clearly showed that pre-treatment with A8301, rapamycin, and resveratrol shifted the metabolome of PD EnSCs in day 3 of decidualization toward that of WD sources represented by co-clustering with WD sources, while untreated PD EnSCs showed a separate pattern of clustering (Fig.  9 A–D, Supp. Data 1). Fig. 9 Metabolomics shift of PD to WD in EnSCs pretreated with senomorphics. PD EnSCs cells were pretreated with senomorphics for 5 days before decidualization and then in vitro decidualized. Hierarchical clustering and Partial least squares discriminant analysis (PLS-DA) for possible effect of ( A ) A8301-1 μM, ( B ) A8301-500 nM, ( C ) rapa (rapamycin) and ( D ) rez (resveratrol) on poor-decidualized (PD) hEnSC at day 3 of decidualization to find metabolome shift toward well-decidualized (WD) cells. The color scale in heatmap represents the relative abundance of metabolites, with red indicating higher abundance, deep blue indicating lower abundance, and light blue indicating zero abundance. The lower side of the figure shows the sample codes, and the dendrogram on the left and top represents the clustering results of the differential metabolites Metabolomics shift of PD to WD in EnSCs pretreated with senomorphics. PD EnSCs cells were pretreated with senomorphics for 5 days before decidualization and then in vitro decidualized. Hierarchical clustering and Partial least squares discriminant analysis (PLS-DA) for possible effect of ( A ) A8301-1 μM, ( B ) A8301-500 nM, ( C ) rapa (rapamycin) and ( D ) rez (resveratrol) on poor-decidualized (PD) hEnSC at day 3 of decidualization to find metabolome shift toward well-decidualized (WD) cells. The color scale in heatmap represents the relative abundance of metabolites, with red indicating higher abundance, deep blue indicating lower abundance, and light blue indicating zero abundance. The lower side of the figure shows the sample codes, and the dendrogram on the left and top represents the clustering results of the differential metabolites

Reproduction is the fundamental outcome of evolutionary processes and is essential for all living organisms to optimize their fitness within their environment. Successful fertility depends on a reciprocal interaction between the embryo and the endometrium, with the health of the endometrium-referred to as the quality of the soil-playing a central role, rather than the seed or embryo itself [ 1 ]. The healthy endometrium responds to the quality of embryo, assessing its suitability for implantation and supporting the progression of pregnancy [ 2 ]. The stromal compartment, populated with a large number of endometrial stem/stromal cells (EnSCs), makes up the largest proportion of the endometrium and controls tissue proliferation, remodeling, and breakdown during the menstrual cycle, under the tight regulation of estrogen and progesterone, and local and peripheral immune cells [ 3 , 4 ]. Since the first report on the immunomodulatory properties of mesenchymal stem cells (MSC) in 2002 [ 5 ], numerous studies have shown significant impact of EnSCs on the function of key cells of the innate and adaptive immune systems including macrophage polarization, dendritic cell maturation, natural killer (NK) cell proliferation and cytotoxicity and T cell proliferation and polarization [ 6 ]. Recent data on single-cell analysis of endometrial cells revealed EnSC subpopulations in pregnant mice endometrium capable of immune cell recruitment, controlling local immune effector mechanisms as well as regulating vascularization, highlighting their pivotal role in maintaining decidual homeostasis. Dysfunction of EnSCs lead to abnormal immune cell accumulation and abortion [ 7 ]. From an evolutionary perspective, EnSCs in menstruating species undergo a spontaneous differentiation process termed decidualization in each menstrual cycle to prepare for a possible pregnancy. Decidualization involves widespread changes in the transcriptome, proteome, and metabolome of EnSCs in both humans and mice [ 8 ]. Although there are many studies focused on the impaired function of immune cells linked to abortion and various pregnancy complications, recent evidence suggests that defective decidualization plays a central role [ 9 , 10 ]. Decidualization failure has been associated with a higher history of miscarriages [ 10 ]. Stromal cells from women with severe preeclampsia exhibited an inability to undergo decidualization, leading to impaired trophoblast invasion [ 11 ]. Additionally, a proportion of patients with recurrent pregnancy loss demonstrated impaired cyclic decidualization and response to hormonal signals resulting in dysregulation of key decidualization markers [ 12 ]. Decidualization imposes a second level of immunomodulatory properties in EnSCs. Accordingly, we showed recently that decidualized EnSCs could normalize abortion rate in CBA/J X DBA/2 abortion-prone mice model by upregulating the frequency of regulatory T cells (Tregs) in uterine draining lymph nodes [ 13 ]. Preventing abortigenic T cell chemotaxis in the decidua [ 14 ], inhibiting uterine NK (uNK) cytoroxicity [ 15 ] and antigen presentation and differentiation of monocytes to mature dendritic cells [ 16 ], and polarization of decidual macrophage to M2 phenotype [ 17 , 18 ] are among the reported immunomodulatory properties of decidualized EnSCs. Recent evidence indicates that decidualization is a multi-stage differentiation process that begins with an evolutionarily conserved acute cellular stress response [ 12 ]. This initial phase is characterized by the appearance of senescent decidual cells, which are able to produce inflammatory mediators, senescence-associated secretory phenotype (SASP), including cytokines such as Interleukin (IL)−6 and IL-8. As decidualization progresses, progesterone-dependent and stress resistant subtype of decidual stromal cells (DSC) dominates that is characterized by the secretion of prolactin (PRL) and insulin-like growth factor binding protein-1 (IGFBP1). This subtype is the dominant cell population that controls the invasion of trophoblast, sense the quality of embryo and supports a successful pregnancy. If decidualization continues and the senescent cell population remains, all endometrial cells may enter a state of chronic senescence. This phenomenon is commonly observed in vitro in the absence of uNKs. However, in the human endometrium, DSCs recruit uNK cells to clear the senescent subpopulation, thereby maintaining decidualization [ 19 ]. Recently, the use of senotherapeutics such as senolytic drugs, which eliminate senescent cells, and senomorphics, which modulate their activity and secretome, has gained attention for enhancing fertility [ 20 ]. Given the critical role of senescent cells in the early stages of decidualization, the timing and dosage of these treatments must be carefully managed. The dominance of DSC subpopulation continues as long as progesterone levels remain within an acceptable range. If implantation does not occur, progesterone level drops leading to recruitment of inflammatory cells to decidua [ 3 , 21 ], secretion of proteolytic enzymes, tissue breakdown and menstruation. Menstrual blood contains fragments of endometrium and menstrual blood stem/stromal cells (MenSCs) [ 6 , 22 ]. MenSCs have long been considered as a non-invasive and readily available counterpart to EnSCs, sharing several key characteristics. Both cell types exhibit mesenchymal markers and share immunomodulatory and therapeutic effects in vitro and in various preclinical animal models. Indeed, we and others demonstrated that both cell types exhibit the ability to differentiate into multiple lineages [ 23 – 26 ]. Nonetheless, the origin of MenSCs remains an active topic of discussion in the field. Based on pathological evidence from our recent study [ 27 ], we proposed that MenSCs are likely derived from the endometrial functional layer, although this requires further investigation. While it is understood that MenSCs are shed from the endometrium and likely represent a mixture of stromal fibroblasts and EnSCs, their precise composition and potential modification by menstruation-associated signaling (such as the inflammatory NF-κB-mediated cascade, Wnt, and Notch pathways) and cell culture conditions warrant further in-depth analysis [ 22 ]. Indeed, they are collected at a different phase of the menstrual cycle, which may result in differences in their properties. Notably, MenSCs undergo hormone withdrawal, potentially compromising their decidualization potential. In this context, we provide the first comprehensive comparison of the decidualization capacity of EnSCs and MenSCs, evaluating key markers of functional decidualization in both. Special focus was placed on metabolomic profiling and pro-inflammatory cytokine production, alongside assessing senescence status and how each cell type responds to senomorphic treatments during decidualization.

Here, we present the first detailed analysis comparing the decidualization potential of EnSCs and MenSCs, examining critical functional markers of this process. Our investigation prioritized metabolomic changes and pro-inflammatory cytokine secretion, while also evaluating cellular senescence and the effects of senomorphic therapies on both cell types during decidualization. This study demonstrated for the first time that EnSCs exhibited a higher decidualization potential than MenSCs represented by the higher PRL expression, circularity index and secretion of PRL at both PC and EPC decidualization protocols. This differential decidualization capacity can be attributed to progesterone-resistant nature of MenSCs. MenSCs have experienced hormone withdrawal and are less responsive to progesterone compared to EnSCs. Hormone withdrawal of EnSCs has shown to be linked with an inflammatory phenotype and production of a host of inflammatory mediators by the decidualized stromal cells [ 40 ]. Of note, the lower decidualization capacity of MenSCs became even more pronounced in the cells remained in culture for longer periods before in vitro decidualization, consistent with the higher susceptibility of MenSCs to senescence than EnSCs. Indeed, MenSCs showed higher baseline and post-decidualization SABG activity, further supporting their predisposition to senescence and impaired decidualization. This assumption aligns with previous findings, suggesting that senescent EnSCs, likely through their altered secretome, induced a ‘bystander’ effect that suppressed the decidual response in neighboring cells, ultimately impairing decidualization and implantation [ 41 ]. Examining metabolome changes during cell differentiation is crucial for a deep understanding of cell biosynthetic demands, regulation of signaling pathways, and complements transcriptome and proteome analyses for identifying specific biomarkers [ 42 ]. With this perspective, we compared EnSCs and MenSCs to assess metabolome shifts during decidualization. Segregation between decidualized and undecidualized cells under PC protocol occurred on day 3 in MenSCs and day 6 in EnSCs indicating that MenSCs have faster but limited decidualization potential compared to EnSCs. This was also the case for the EPC protocol, in which decidualized MenSCs produced PRL as early as 3 days after decidualization, while for EnSCs significant difference was observed on day 6. Another notable finding was the identification of shared pathways among MenSC day 3 and EnSC days 3 and 6, distinct from MenSC day 6. These conditions exhibited increased levels of phenylalanine, tyrosine, and methionine, along with decreased citrulline. Previous studies have reported elevated phenylalanine and tyrosine during decidualization [ 43 ]. Moreover, upregulation of phenylalanine and tyrosine metabolism is observed in the dot plot of shared pathway enrichment in MenSC day 3 and EnSC days 3 and 6, and this pathway has previously shown to be upregulated during decidualization [ 44 ]. Additionally, the methionine cycle, which generates S-adenosylmethionine (SAM) as a methyl donor for gene expression regulation, is critical for decidualization, as inadequate methylation disrupts this process [ 45 , 46 ]. Citrulline, a precursor to nitric oxide (NO), an essential regulator of decidualization that is produced under the influence of progesterone during the luteal phase [ 47 ], was also observed to be decreased. Collectively, our data indicate that MenSCs have distinct decidualization kinetics compared to EnSCs with faster, limited and sustained capacity to decidualize for an extended period of time. There are plenty of evidence reinforcing the pivotal role of glucose metabolism during decidualization. The general consensus is that decidualization is a glucose-dependent process, in which glucose uptake significantly increase by decidualization through different mechanisms including activation of insulin signaling-related genes and the expression of glucose transporter, GLUT1 , in decidualized EnSCs [ 48 , 49 ]. Low glucose inhibits decidualization by decreasing FOXO1 expression [ 50 ]. However, the aforesaid conclusion is mostly based on the investigation of glucose metabolism regulatory pathways. There is very limited number of reports that have directly measured glucose in cell culture supernatant of decidualized EnSCs. NMR-based characterization of metabolites extracted from the culture medium showed that decidualization of EnSCs is associated with decreased levels of glucose [ 43 ]. This finding is in sharp contrast with our finding showing that decidualization in EnSCs is linked to a notable increase in glucose concentration a concomitant decrease in lactate production in cell culture supernatant. We think that the basis for this disparity is due to the cAMP concentration in decidualization protocol. We used PC protocol that comprises 10 times more cAMP compared to the EPC protocol. Notably, high concentration of cAMP causes a sharp decrease in GLUT1 expression leading to decreased glucose uptake [ 51 ]. It is worth noting that the previous observation about decrease in glucose levels during decidualization is reasonable, given that 50 μM cAMP is used to induce decidualization in EnSC under those conditions [ 43 ]. Decidualization is mainly associated with a cease of cell proliferation, while undifferentiated cells continue to proliferate, which may explain lower consumption of glucose in decidualized cells. Notably, senescent cells consume more glucose [ 52 ] by glycolytic pathway, known as Warburg effect, which may explain comparable level of glucose consumption in decidualized and undecidualized MenSCs. There is compelling evidence that impaired decidualization disrupts embryo-maternal interactions and causes recurrent pregnancy loss [ 53 – 55 ]. In this context, PD and WD EnSCs and MenSC was compared to explore the mechanistic link between poor decidualization and senescence, metabolic activity and responsiveness to senomorphics. We stratified cell sources to poor and well decidualized based on the median level of PRL secretion [ 32 , 56 ]. To validate this approach, we performed an unbiased analysis using the median absolute deviation (MAD) to calculate robust z-scores. This method confirmed that positive z-scores correspond to WD sources and negative z-scores to PD sources, validating our initial stratification. Indeed, unbiased analyses including metabolomic profiling, decidualization kinetics, response to senomorphics, and senescence patterns consistently clustered into distinct groups corresponding to WD and PD phenotypes. This independent validation confirmed that a PRL median cutoff is a justifiable and suitable threshold for stratification of WD and PD cell sources. Importantly, it was reported that high PRL-expressing EnSCs co-expressed key decidual markers like IGFBP1 , FOXO1 , and SCARA5 and more effectively supported BeWo spheroid invasion confirming PRL as a valid indicator of a robust decidualization phenotype [ 41 ]. PD cells exhibited greater senescence compared to WD cells, with significant differences in EnSCs under the PC protocol and in MenSCs under the EPC protocol. It seems that in an aging cell population, the use of stressors at a lower concentration for a longer time can better reveal the difference between cells with different senescent status. This is the case for PD and WD MenSCs differentiated for 12 days with EPC protocol, which contains ten times less concentration of cAMP. Our results also highlighted the contrasting effects of senomorphic pretreatment on the senescence of PD and WD EnSCs and MenSCs. Pretreatment with all senomorphic agents decreased senescence in PD EnSCs but induced senescence in WD sources suggesting that the effects of senomorphics is differentially regulated depending on the cellular initial stress level. As shown earlier, PD sources exhibited higher SABG activity compared to WD sources and contained higher proportion of senescent cells. It is conceivable to imagine that pre-treatment of PD EnSCs with senomorphics decreased the frequency of senescent cells and counteracted with the excessive production of SASP. This assumption is in line with the diminished levels of IL-6 in PD EnSCs following pre-treatment with senomorphics. The immediate question that arises is why WD EnSCs responded in opposite manner to senomorphics. It seems that pathways activated in low-senescent WD EnSCs following senomorphics treatment are already saturated in high-senescent PDs. For instance, resveratrol, a well-known senomorphic agent, exhibits distinct effects on senescence induction in normal versus senescent cells. This difference is partly due to the activation of distinct pathways or the differential regulation of the same signaling pathway in each cell type during senomorphic treatment [ 57 ]. Additionally, the contrasting effects of resveratrol can be linked to variations in the redox status of the cells [ 58 ]. A similar dichotomy has been observed in the regulation of the Nrf2 stress-responsive signaling pathway following polyphenol treatment, as the downstream targets of Nrf2 differ between cell types, highlighting cell-specific responses [ 57 ]. It has been demonstrated that p53 can both induce and suppress senescence based on the status of cell cycle arrest and the expression of p21 in cells [ 59 ]. Moreover, like other cells in the body, senescent cells are heterogeneous population, and their response to treatments depends on subpopulation-specific functions. Cells arrested in G2, which exhibit higher senescence and secrete more IL-6, are more sensitive to senolytic agents than G1-arrested cells with lower senescence [ 60 ]. Obviously, senomorphic agents do not eliminate senescent cells, and differences in response are evident at the level of markers like SA-β-Gal and IL-6. Consistent with these findings, we found that PD cells, whether from EnSCs or MenSCs, responded to senomorphics by decreasing IL-6 production. This underscores that the impact of senomorphics depends on the stage of senescence and associated cellular pathways. Interestingly and consistent with our finding in that PC protocol was unable to distinguish between senescence of PD and WD MenSCs, differential responsiveness of WD and PD EnSCs to senomorphics was not observed in MenSCs suggesting that MenSCs could not properly show the effect of senomorphics on cellular senescence. Another interesting observation was the differential impact of senomorphics depended on treatment timing. When the cells were treated with senomorphics at the onset of decidualization, they induced senescence in EnSCs and MenSCs regardless of their decidualization capacity, except for PD MenSCs, which were unresponsiveness to this treatment. From these findings, two practical points of view can be inferred. First, the use of senomorphics should not routinely be ordered for women with infertility or recurrent pregnancy loss, rather it should be tailored to patients with suspected endometrial senescence and impaired decidualization. It should be noted that from the drugs examined in the study, metformin and prednisolone are utilized in IVF protocols for specific groups, such as patients with polycystic ovary syndrome (PCOS) or recurrent implantation failure (RIF), to manage conditions like insulin resistance or immune dysregulation [ 61 , 62 ]. It was noteworthy that pretreatment with senomorphics could shift the metabolome of PD sources toward WD EnSCs suggesting their beneficial outcome in women with poor decidualization and reproductive failure. This is consistent with recent findings that exposing primary cultures to senomorphics at the onset of decidualization prevents inflammatory reprogramming and specialized decidual cell formation. However, pretreatment does not impair decidualization, highlighting the importance of optimizing the dose and timing of senomorphic exposure for desired outcomes [ 63 ]. Second, administering senomorphics in embryo transfer cycle could have a negative effect on endometrial senescence and receptivity. At the glucose metabolism level, PD and WD EnSCs and MenSCs behaved differentially, which could be attributed again the senescent nature of MenSCs especially in PD sources. Concerning the impact of senomorphics, a reduced rate of glucose utilization was observed exclusively in PD EnSCs, consistent with the rejuvenation activity of senomorphics over PD EnSCs and resulting lower glucose consumption. Although EnSCs and MenSCs showed differential pattern of decidualization, senescence and response to senomorphics, the ideal scenario would have involved collecting samples from a single individual, which in most cased was impossible. The history of infertility in the EnSC group may serve as a confounding factor. Indeed, higher number of donors, especially for EnSC group, could increase the robustness of results. Although our study primarily focused on the characterization of decidualization dynamics in MenSCs and EnSCs, we acknowledge that exploring the molecular mechanisms involved including differential expression of estrogen (ER) and progesterone receptor (PR) [ 64 , 65 ] would provide valuable insights. Another limitation is that while recent studies have demonstrated that our decidualization induction method closely mirrors transcriptomic changes seen in in vivo decidualization, it is still unclear whether these results entirely represent the in vivo context [ 66 ]. Indeed, there is no standardized protocol for distinguishing between WD and PD groups and lower PRL levels do not necessarily indicate impaired decidualization. Further research is warranted to reveal differences in the other biological aspects of MenSCs and EnSCs.

In summary, our data reveal significant functional distinctions between MenSCs and EnSCs, especially in their hormonal responsiveness. Although MenSCs remain a convenient and accessible cell source, our results indicate that they may not accurately replicate EnSC biology or serve as an optimal replacement for EnSCs in endometrial research—particularly in studies investigating decidualization or screening senomorphic compounds. This limitation warrants careful consideration, especially in clinical trials aiming to utilize MenSCs to correct impaired decidualization. Future studies employing MenSCs in place of EnSCs should therefore be conducted with caution. These distinctions underscore the importance of selecting appropriate cellular models in reproductive research and may inform future strategies for fertility-related drug development and personalized medicine. This study also highlighted the inherent differences of WD and PD EnSCs and MenSCs especially their different responses to senomorphics and recommended against routine use of senomorphics in infertile women especially at around the window of implantation period.

Supplementary Figure S1. Classification of EnSC and MenSC cell sources by robust Z-score of PRL secretion. Lollipop plot displaying the robust Z-score calculated from prolactinsecretion levels was indicated in EnSC and MenSC cell sources, applied as a classification metric. Positive Z-scores indicate PRL secretion above the reference mean), while negative Z-scores indicate secretion below the mean).) Supplementary Figure S2. Assessment of cell senescence in EnSCs and MenSCs following decidualization. Cells were decidualized by EPC protocol and SABG activity in undecidualized and decidualized EnSCsand MenSCswas measured and compared. Panelanddepict representative SABG staining and activity in well-decidualizedand poor-decidualizedcell sources, respectively.Level of IL-6 in cell culture supernatant of undecidualized and decidualized EnSCs and MenSCs at different time points of decidualization. Each dot is attributed to single cell source. P values less than 0.05, 0.01, 0.001 and 0.0001 are shown with *, **, ***, **** respectively. IDV, Integrated density value.) Supplementary Figure S3. Correlation of PRL secretion with IL-6 production and senescence. EnSCs and MenSCs were in vitro decidualized by PC protocol and at the end of differentiation process, levels of PRL, SABG activityand IL-6were measured. The same setting was done for WD and PD cells. Spearman correlation was done to assess potential correlation between PRL secretion and IL-6 production or SABG activity. Scatter plots with best fit lineand 95% confidence intervalare shown.) Supplementary Data 1 Supplementary Table 1 Supplementary Table 2 Supplementary Figure S1. Classification of EnSC and MenSC cell sources by robust Z-score of PRL secretion. Lollipop plot displaying the robust Z-score calculated from prolactinsecretion levels was indicated in EnSC and MenSC cell sources, applied as a classification metric. Positive Z-scores indicate PRL secretion above the reference mean), while negative Z-scores indicate secretion below the mean).) Supplementary Figure S2. Assessment of cell senescence in EnSCs and MenSCs following decidualization. Cells were decidualized by EPC protocol and SABG activity in undecidualized and decidualized EnSCsand MenSCswas measured and compared. Panelanddepict representative SABG staining and activity in well-decidualizedand poor-decidualizedcell sources, respectively.Level of IL-6 in cell culture supernatant of undecidualized and decidualized EnSCs and MenSCs at different time points of decidualization. Each dot is attributed to single cell source. P values less than 0.05, 0.01, 0.001 and 0.0001 are shown with *, **, ***, **** respectively. IDV, Integrated density value.) Supplementary Figure S3. Correlation of PRL secretion with IL-6 production and senescence. EnSCs and MenSCs were in vitro decidualized by PC protocol and at the end of differentiation process, levels of PRL, SABG activityand IL-6were measured. The same setting was done for WD and PD cells. Spearman correlation was done to assess potential correlation between PRL secretion and IL-6 production or SABG activity. Scatter plots with best fit lineand 95% confidence intervalare shown.) Supplementary Data 1 Supplementary Table 1 Supplementary Table 2